Featured Post

Election 2016: Last Call for Predictions

Alright, we’ve all talked about what we think is going to happen. We’ve all made predictions. Some very specific predictions, some changing predictions, and some wide-open predictions. But the election is tomorrow, and it’s time to lay down our markers.

What do you think is going to happen? What will the national popular vote be? Who will win what state? What odd specifics do you think are going to happen?

My prediction has been pretty constant since the convention at least, which is Clinton+6 and about 350 electoral votes. Here is the map, but the long and short of it is that I have Clinton winning every non-red swing state except Iowa and I have her picking up Arizona. For the popular vote, I’ll go with: Clinton 51%, Trump 45%, Johnson 3%, Other 1%. My off-the-wall prediction is that we’re going to find out that Clinton won North Carolina before we find out that Trump won Georgia. Race will be called pretty early, though not sure when.

In the senate, I predict the GOP wins Arizona, Indiana, Ohio, and Florida and lose Illinois, Wisconsin, Pennsylvania, New Hampshire, Nevada, and Missouri. This is a 52-48 Democratic majority.

I predict the GOP will lose 12 House seats.

Feature Image


Editor-in-Chief
Home Page Twitter Google+ Pinterest 

Will Truman is a former professional gearhead who is presently a stay-at-home father in the Mountain East. He has moved around frequently, having lived in six places since 2003, ranging from rural outposts to major metropolitan areas. He also writes fiction, when he finds the time. ...more →

Please do be so kind as to share this post.
TwitterFacebookRedditEmailPrintFriendlyMore options

195 thoughts on “Election 2016: Last Call for Predictions

  1. Clinton 49.2 Trump 44.9 Johnson 4.8 Stein 0.8%, everyone else combined 0.2%
    Electoral College Clinton 328 Trump 210 http://www.270towin.com/maps/VgBZ

    GOP retains House, Senate 50/50, giving Dems control by virtue of Kaine.

    I’m still liking what I said earlier. I might have underestimated Stein by a point (which are lost Clinton votes) and maybe overestimated Johnson (which are gained Trump votes). It also looks like link rot got to my electoral college map, and I can’t remember the exact states.

    this might have been what it was, the electoral vote numbers are the same.

    Report

    • Here’s my US Senate map. My electoral college map is probably wrong above wrt New Hampshire, because there’s no way Trump wins the state and Ayotte doesn’t win her re-election – and right now, things are breaking in a way that both Trump and Ayotte lose in NH. So my final electoral college prediction is this map, with Clinton 322, Trump 216.

      Report

  2. In the previous thread, I think I overestimated the third party vote share and somewhat underestimated Clinton’s EC performance. Final prediction: Clinton 50, Trump 45, Johnson 4, others 1; Clinton 341-197 in the EC, on this map.

    Report

    • Your map, , looks similar to mine (my more current prediction) below. Differences are Maine-2, North Carolina, and Ohio. You say Clinton carries them, I say Trump does.

      I notice we both think she takes Florida, which is another very close call. But note that even on my more pessimistic (which is to equate pessimism with a strong showing by Trump) map, flipping Florida to the GOP still results in a Clinton win, albeit a squeaker rather than a triumph.

      Report

  3. Today’s cinematic scenario:
    Trump wins, 50%-48% (Yes, there’s a Civil War in the Powers that Be, this is spillover).
    Trump pardons Clinton (Trump is indeed too stupid to see this as a problem).
    A few days later, someone assassinates Trump.[Q: Left or Right? A: Rightwing Loony].
    Say hello to President Pence.

    [Given that Trump appears to have gotten some of his campaign ideas from Doctor Who (someone sent him their notes), this may not be as farfetched as you might think.]

    Report

    • Even if Trump is elected — a result which will make me very very unhappy — I would not want him to be assassinated. Period, full stop. Some kinds of national concerns must, and do, transcend partisanship or other forms of political distaste for particular officeholders.

      I would want his Presidency to be frustrating, I would want his policies to be obstructed and his appointments kept in strong check by an independent and politically counterbalanced Senate (and maybe even a House), and I would want to see him as restrained by aides as much as possible when making public speeches. I would wish for him to be a singularly ineffective President and for him to be voted out of office in 2020. Were and when we to go to war under a President Trump, I would hope (faintly) for a swift victory even knowing that he would claim credit for it.

      An assassination would do the whole country tremendous damage. An impeachment, too, although different kinds of damage and in different ways. Better to muddle through with bad people in office and await the results of the next election cycle.

      Report

          • That sounds more like it. Trump wouldn’t be interested in doing the day to drudge work, so he’d appoint people from his clearly elite squad of trusted advisers to do it. This should not make us feel better.

            And if Trump won, he’d certainly have Republicans in the House and Senate. I’m not sure how the party that couldn’t keep him in check during the primaries and election would go about keeping him in check once he was actually sitting in the Oval Office. At least a Democratic Congress could oppose his agenda without losing votes from their constituents. A Republican Congress would quail the first time he railed against them in a press conference for not getting things done.

            Report

            • tf,
              Easily. The donors want “not trump” and you’d get at worst a detente and at best a recession in Imperial Presidency Power.

              We have secret laws on the books, now. You can’t even read them without a security clearance.

              It is worth it, at this point, to elect a REALLY unpopular president, just to have him cede some authority to Everyone Else In Washington

              Report

          • Trump was bought off. Then he proved so inept at losing that they had to bring out the big guns (all that stuff about him being a mysogynistic ahole).
            You should listen to Trump apologizing for proving too likeable.

            Report

      • Burt,
        This is a prediction, a malediction even, not a “I want this to happen.”
        I consider it marginally more likely than Trump taking office.
        This is what happens when your election becomes part of an ongoing Civil War.

        Report

          • In my bubble, the Powers that Be have lost multiple things that they expected to go their way this year (ya might remember Brexit). It may be that they’re not going to chance this one turning out the way the voters want.
            [And if you think that the Powers that Be DONT have the capability to turn an election on a dime, you’re the one in the bubble not me].

            Report

              • 1) They vastly overestimated their PR.
                2) At some point, fixing the vote totals becomes obvious. You can pretty much get away with a 3% swing, or even a 5%, but any more than that, and you’re pretty much at the “warning, do not do this” level.
                3) They got surprised.

                The Powers that Be have both Clinton and Trump in the bag, so they’re less likely to care about this than Brexit, which they really, really didn’t want.

                Report

    • [Given that Trump appears to have gotten some of his campaign ideas from Doctor Who (someone sent him their notes), this may not be as farfetched as you might think.]

      “Someone”, Kim?

      Someone“?

      Now Trump’s entire campaign is starting to make sense.

      Report

  4. I think it will be a plurality election with Clinton winning between 48 or 49 percent of the vote.

    Looks like Latino voting is going to help Clinton in Nevada, Florida, and North Carolina. Maybe Arizona and Georgia but probably not.

    Wild cards: Patrick Murphy picks up a surprise victory because of record Latino turnout in Florida. So does Ted Strickland for Ohio. Burr losses NC if HRC wins the state.

    Real wild cards: HRC wins Georgia and Texas

    Report

      • Although this doesn’t look like the year Texas flips, its not outside the realm of possibility.
        These polls at PEC show that Team Red is pulling just under 50%, with Team Blue around 40%.
        A surge in Latino votes not picked up by pollsters, lack of a ground game by Trump, and these numbers could swing a few points.

        Not enough to flip, this year. But 2020 is looking pretty good.

        Report

        • I know these things! And I’m fascinated to see how the slow gain of demographic creep will play out against the straightjacket of overt gerrymandering there.

          But this isn’t going to be the year that the state as a whole flips and legislative control gets out of sync with the statewide popular vote. An 8-10% differential means that Texas is still deep red.

          Report

          • Depends on where the losses are. Texas has a lot of educated whites who vote Republican. If they vote Democratic, and if he loses too many Hispanics (and inspires them to vote against him) we could see a much larger swing there than we see nationally.

            There are reasons Texas and Georgia are polling more closely than they should, I can’t remember who pointed it out, but the Republican coalition is moving north and the Democratic one is moving south. Maybe. It could be that tomorrow, educated whites and men and women end up roughly where they were in 2012.

            (I should note that I’ve long been a skeptic of Blue Texas. For all of the demographic changes, it just hasn’t budged. It’s the revolution that never seems to come. But if there’s a reshuffling of the electorate, that would redound to the benefit of Democrats far more than the creep of demographic change that hasn’t.)

            Report

      • I think Clinton loses Texas by about half of what Obama lost Texas by, and that half of that change is permanent — not uniquely anti-Trump, so much as ‘solidified Democrat’. (New voters, mostly, and some people who finally said “Screw it”. Texas is pretty much a one-party state at the top, so when the tipping point finally comes — the collapse will be epic. But that’s a decade off, minimum).

        Report

      • I said it was a wild card!

        I do think NC will go blue. A judge restored thousands of purged voters. The DOJ is sending monitors. The Latino vote is surging there and did early voting.

        Report

            • While speaking with Obama on Friday, Mitú’s Gina Rodriguez said: “Many of the millennials, DREAMers, undocumented citizens — and I call them citizens because they contribute to this country — are fearful of voting.”

              She then asked: “So, if I vote, will immigration [U.S. Immigration and Customs Enforcement] know where I live? Will they come for my family and deport us?”

              “Not true,” Obama responded. “First of all, when you vote, you are a citizen yourself, and there is not a situation where the voting rolls are transferred over and people start investigating, et cetera.”

              “The sanctity of the vote is strictly confidential in terms of who you voted for,” Obama said.

              Illegals shouldn’t fear voting is the only message I see. Otherwise why even mention them at all? Obama seemed perfectly fine with the idea of illegals voting.

              Report

                  • As Mike pointed out to you already, reading comprehension could use work.

                    That very clearly means:

                    “If you are voting, it means you must already be a citizen. We are not sending lists of citizens with Latin names to immigration officials so they can check whether there are non-citizens living alongside citizens who have voted.”

                    I don’t think you’re stupid, notme – I think you’re reasonably intelligent and are only able to sustain this level of incomprehension through a sustained effort at suppressing your own intelligence.

                    Report

        • As many of the other forecasting pundits have written, the USC tracking poll is an interesting experiment that tries to measure things in different ways. It’s not surprising that some of the assumptions built into a new model are wrong and result in a systematic bias. Eg, Nate Silver points out that when you ask people who they voted for in the previous Presidential election, as the USC poll does, there’s a significant number of people who lie/misremember in favor of the winner.

          The map is a separate opinion of several of their writers who consider all sorts of information beside the USC poll.

          Report

          • Ernie Tadeschi (@ernietadeschi on Twitter) has done the experiment of reweighing the USC poll based on US census data without paying attention to who people said they voted for last time; doing so has lead to results that look a lot more like those of other polls.

            It’s not a slam dunk argument that this is the right way to do things, but I think it does at least wink suggestively in that direction.

            Report

    • A number of red states will have surprising turnout for Clinton, but stay VERY red on the down-ticket races. She will top 50% in National popular vote by mere fractions.

      GA will be close, but parts of Texas will go into recount from GOP challenges, and heads will explode.

      Report

  5. I’m going to side with you Will. There will be an overwhelming Clinton in the electoral college and a plurality win the popular vote. Democrats will take the Senate by a small majority in the Senate. The House will remain Republican.

    Report

    • Seconded. The landslide hopes were killed when Comey stuck his oar in and now turns out the whole collection was a nothingburger. I hope neither Obama nor Clinton try to fire him. I also hope he has the decency to resign.

      Report

      • I think you mean either, not neither. Comey’s actions is not only a breach of a very big norm, it was arguably illegal rather than an act of constitutional hardball.

        Report

        • For what it’s worth, I have an acquaintance (he’s a friend of my brother’s) who works in the FBI. (Not a field agent, analysis side). My brother asked about the whole Comey thing when it happened, and he said something like “I’ve never seen anyone try that hard to get fired”.

          This was, IIRC, before the OTHER set of leaks implicating people in the NY field office (former and current) that seemed to be really good friends of Rudy Guiliani, and the implications that had.

          He might have a different take now, but his first take after the letter was leaked was that Comey wasn’t courting firing, he was jumping up and down and demanding it.

          (For what it’s worth, this guy is…really conservative. Probably not a Trump voter, but no fan of Clinton at all. He’s also Got Opinions about field agents in general. I got the distinct impression that the field agents and the analysis folks do not always see eye to eye).

          Report

        • Going after Comey would have looked like petty partisan politics as long as the investigation remained open. Now that Comey is back on the list of people who are part of the Grand Clinton Conspiracy, it seems like it would be more politically feasible.

          Report

          • tf,
            People pulled him OFF that list? Idiots and fools, all of them.
            Clinton understands one thing: Too Big To Fail.
            Her ULTIMATE BACKUP PLAN?
            Buying all the electors in the Electoral College.

            No, I am not joking, this is straight from someone reasonably high ranked in Clinton’s team.

            Report

      • The landslide hopes were killed when Comey stuck his oar in

        No. The landslide hopes were killed when the Dems decided it was good politics to give the nomination to a fair-at-best politician who wasn’t particularly likable or popular.

        Report

        • Not this again. Clinton won the Democratic nomination for the same reason that Trump one the Republican nomination, she was the most popular with the majority of the people involved in intra-Democratic politics. Many more people supported her than Bernie Sanders, her nearest rival. Everybody else who entered the Democratic primary had no chances but unlike the Republican primary, clearly realized it and dropped out. Even assuming everything you believe about Hillary Clinton is correct, who else could the Democratic Party nominate in her stead? She was the most nationally well known politician in the Democratic Party not named Obama who wanted the job. Warren wanted to remain in the Senate. Biden wanted to retire for a variety of reasons.

          I also think that the constant fake scandals that the press drummed up against Hillary Clinton since the 1990s is the cause of a lot of the hatred against her. She was a relatively popular Senator and Secretary of State. Clinton surged in popularity whenever the press was not on her back and had her polls drop when the press was on her back. Donald Trump does better when he is not really in the national spotlight and fails when he can access his Twitter account.

          Report

          • I think you’ve missed the point. Nothing you say here changes the dynamics that made it especially hard for Clinton to win by a large margin compared to another candidate. So not “this again” but rather “still this.”

            Report

          • My impression has been that Sanders achieved the goal he set out to: to move the Overton window to the left. He didn’t drop out of the race because the actual nomination was a very distant stretch goal to him.

            The other Democratic candidates for nomination dropped out because their goals were to win, with any considerations of effecting political change through the primary campaign itself distant to nonexistent.

            Report

            • He’s also a national star now who can lay his hands upon state-level political issues all over the place and offer unto them his Bernie Sanders blessing, thus giving them a bump among young liberal voters. I don’t think he had that super power before the primary.

              Report

          • Hillary’s success is more do to winning the hidden primary before any votes were cast. That is where she gained enough institutional support to clear the deck of all credible opponents. I say this not as a criticism of her, but rather that she largely won before a single primary was held. I am convinced that this was 90% of Bernie’s popularity. If someone like Hickenlooper or Biden jumped in, she would have been less likely to win.

            Report

              • Eh, I like Feingold, but he’s very popular with upper middle class white people who have very different priorities than the rest of the party. He’d have the same problem that Bernie had – the actual base of the Democratic Party is working class African American and Hispanic voters based on transactional politicians with long established ties, not inspirational figures. These voters don’t really care about the Patriot Act, and Iraq War vote from 10 years previous, or campaign finance reform.

                Remember, the black vote didn’t shift to Obama until _after_ he beat Clinton in Iowa.

                Report

          • I love when you guys argue things that are essentially regional differences :^)

            Outside of their home state and Senate leadership, there’s no such thing as a “popular” Senator. Most voters are hard pressed to tell you who their two US Senators are most of the time. Perhaps in neighboring states in the Northeast, where metro areas are much more likely to cross state lines. (Quick, who’s the junior Senator from Oregon?)

            Ditto for Secretary of State, mostly viewed as the President’s top paid arm twister, not a source for foreign policy.

            Look at the primary maps that show the Clinton-Sanders split (this one, say). She did extraordinarily well in the Deep South (not that that will bring many EC votes with it), moderately well in the Rust Belt and California, and skimmed by in the Southwest. There’s a big chunk of the country from the upper Midwest across to the Pacific (including northern Nevada and California above the SF metro area) where Bernie kicked her butt.

            The Dems are running a NE urban corridor ticket this year. Blue states in the rest of the country will vote for that ticket, but they weren’t ever going to be enthusiastic about it.

            Report

          • Lee,
            When Clinton forcibly removed Bernie Sanders voters from the election, I don’t think you get to say that she “won” the vote for being the most popular.

            P.S. No, this is not a reason NOT to vote for her. Didn’t you say you liked someone like LBJ?

            Bernie would be winning with a larger margin. Even Republicans like Bernie — they say “At least he’s honest” (And I’ve got that out of a pollster’s mouth).

            Report

            • Yup, she’s been up by about 6-8 points since the Convention, with most of the differences being shown by Republicans or Democrats simply not answering polls due to being depressed about the news, not the actual opinions of 300 million Americans changing all that much.

              Ironically, I think the big thing that has actually shifted the most votes is Gary Johnson screwing up foreign policy questions and likely pushing a lot of Johnson voters back to Trump, Clinton, or not voting. It might’ve cost the Libertarians millions of dollars in public funding.

              Report

            • This. Forecasters who take other factors (eg, the economy, historical state voting trends) into account have been around four percentage points in the national vote count and 300-320 EC votes all along. That the median polls have converged to about that here at the end is, in some ways, disturbing.

              Report

              • I’m an icky anti-science poll skeptic, but I can actually think of a benign explanation. If one of the main differences in polling is response rates, than two weeks out people who are winning are more enthusiastic than people who aren’t. However, in the final days of the campaign, everyone is checked in and ready to go. And so things start to line up with other forecast models that don’t depend on response rates.

                Report

        • “Give”?

          I seem to recall an election, with a rather lot of voters, that she won by more than three million votes. With a proportional delegation system designed, in fact, to allow lesser known candidates to stay in the game long enough to really compete, instead of it all being decided in Iowa and NH.

          That’s a funny way to “give” someone something.

          Who did the “giving”, actually?

          Report

  6. Offhand, I’d say Clinton beats her polls in swing states by an average of 2 points (possibly 3, but more likely 2), and nationwide by about a point and a half, and wins by about 5 points. Call it 49-50/44-45.

    Trump will claim this is proof it’s “rigged”. (Everyone else will take this as proof that, indeed, modern GOTV operations are far superior to rallies+commercials).

    Senate goes 51-49.

    It’s been a 3 to 4 point race the entire time, and I think her massive GOTV edge will bump that a point from polls.

    Report

  7. I almost certainly overestimated the third-party effects, but otherwise feel good about what I said last time. Looking forward to 2018, the Republicans make up most of their Congressional losses and continue picking away at the state level. Clinton’s four years are going to look, in hindsight, like a giant game of whack-a-mole, trying to slap down one state after another domestically and one perceived threat after another internationally.

    Report

  8. I suspect that Clinton will win, but it will be something like 49-48 with 3% going to the nutball parties. It’ll be one of those elections where we’ll point to a single state that came close to going Trump and say “man, if this state went 48-49 instead of 49-48, we’d have a President Trump now!”

    There will be two or three such states.

    For some reason, this will be sold as being a “mandate”.

    Report

  9. Way back when, I said this:

    http://www.270towin.com/maps/DAp1n
    Popular vote: Clinton c. 51%, Trump c. 43%, Johnson c. 5%. See further.
    To be followed by much wailing and gnashing of teeth.

    Clinton looks much less strong now, or perhaps it’s that Trump looks stronger although if so it’s in spite of the candidate and because of external events (Comey) and professionals (Kellyanne Conway taking Donald’s twitter away from him). There is also the possibility that the once-commanding Clinton lead was illusory, and what we’re seeing now is a reversion to something resembling even partisan balance that is really based in matters that transcend any candidate or set of candidates.

    Now, it still seems to me that Clinton can play the map. And the basic map still looks good for her, notwithstanding the hits she’s taken recently. So knowing what I know now, reading the polls as I see them know, here are my predictions:

    1. Presidency: Clinton 307 EV and 49% PV, Trump 231 EV and 46% PV.
    1a. Johnson gets about 3%.
    1b. McMullin comes in second in Utah but Trump still wins by a plurality. Not that it matters with Clinton taking a majority of EC votes.
    1c. A few notes about my predictive map: North Carolina goes GOP thanks to the purge of voter rolls there, which results in DoJ litigation against the state that lasts years because, after all, Republican lawmakers targeted African-American voters to make that happen. Substantial but ultimately we’ll-never-know-for-sure evidence will emerge that North Carolina would have voted Clinton. Also, I am still at something of a loss for how Ohio would vote Republican but Pennsylvania would vote Democratic, but that’s the way the polls have been going for more than a month now. Arizona will be the closest election.

    2. The Senate splits 50-50, giving the Democrats nominal control of the Senate by virtue of the Vice Presidency. (Agree with here).

    3. Democrats have a net gain of 10 seats in the House, which is considered a disappointment for them and enough of a stoploss that Paul Ryan maintains the Speakership.

    4. Republicans will fail to learn any meaningful lessons at all from the election and are doomed to ultimately futile attempts to reconcile Trumpian nationalism, now-exposed white privilegists, supply-siders, and religious socons. No one in the party will have the guts to say “We can win without the most obnoxious of these faction members” and initiate a 1968-style purge.
    4a. However, many Trump supporters will become embittered and leave the party; a rump of them will try to form a third party, which efforts will ultimately collapse.
    4b. Meanwhile, structural changes at the RNC level will initiate something roughly analagous to the Democrats’ superdelegate system.

    5. The lame duck session of the Senate will decide that Merrick Garland isn’t so bad after all and make noises about confirming him to SCOTUS but the Republicans will be so busy hating each other it won’t get done.

    Report

    • I think that’s about right. I hate to admit it but I think you’re right, in particular, about North Carolina. It’s not looking good there. Thanks, Roberts Court. Thanks, neo confederates.

      Having spent an awful lot of time in western Pennsylvania and Ohio, I would say Ohio is a lot more racist than PA.

      Report

    • Point four is the rub. I made this point in the past but the Republican Party is having the same problems that the Labour Party in the United Kingdom had between 1979 and 1997 and is having now under Jeremy Corbyn. The party membership in both parties really believe in the party platform but the general electorate of both countries really isn’t buying the platform anymore. The American political system gives Republicans more protection and a greater ability to deny this reality than Labour does though.

      On five, I’m skeptical. Mitch McConnell has stuck firm on his total war oppositional platform. He has vowed that if the Republicans keep the Senate that Clinton will not get any of her appointments through either. The Supreme Court and other Federal courts is something that many religious socons really care about, more than any other segment in American society. The one reason why they are sticking with Trump is to get somebody they like on the Supreme Court.

      Report

    • Hmm, I’m about at this level of pessimism myself, with my map giving her AZ but not NH (I basically assumed any tossup with marginal minority vote will go Trump):

      http://www.270towin.com/maps/dN9l0

      Trump outperforming the Romney map is, in my view, a huge disappointment and should be a big wake up call for the Democrats to start coming up with a coherent, affirmative vision for voters. And, by the same token, if Trump underperforms than Obama will have established both an agenda and a campaign operation that can elect a tire iron if it has a (D) after the name. The party has no obligation screw with something that ain’t broke, and should focus on getting Obama’s face on some currency or a mountain or something.

      Since we’re making long bets, what do folks think about the future of TrumpTV and FOX News? I say TrumpTV starts up but flounders because you just can’t run a whole network with that surrogate talent pool (and his tech talent is looking worse and worse by the day). At FOX: Megyn Kelly leaves; Hannity stays, and the network continues to mirror the party in having no idea what coherent position to take on Trumpism.

      Report

      • Here is where I see TrumpTV having a problem: it needs advertisers. You can finance a website hawking gold and Weird Little Tricks, but not a TV network. Thinking of all the sponsors who pulled out of the convention, and how enthusiastic they’ll be about a network centered around the most controversial part.

        And Trump or Kushner is going to have to convince lenders that it’s viable.

        Something called TrumpTV might happen, but not what people are talking about.

        Report

        • This is a good point. I assumed that there would be enough gold coins, Ronald Reagan commemorate plates, identity theft protection, arthritis medication, and various Trump-brand cross marketing to keep the whole thing afloat. Do you think the market dynamics that have Limbaugh and EIB network signing $400 million contracts would not work for Trump and Kushner?

          Report

      • Yeah advertisers and investors would be an unsurmountable problem for Trump TV. His brand is crap now. Fighting for viewers with Fox would hurt them both but where else does he get them from. His rep is for building stuff and going away to let other people run it. A network isn’t a building type of development so what does he offer except for a losing brand and an older demographic advertisers don’t’ prioritize with an already existing network to reach them. Unless a bunch of rich Russians pony up the bucks nobody will bankroll it and if people were willing it would be only if Trumpy was just the mouth piece and not running it. That doesn’t sound like his kind of gig. Just getting on cable plans would be an up hill struggle. Maybe a Palin style web TV thing which possible.

        Report

        • A couple of points in Trump’s defense:

          * Jared Kushner runs the New York Observer, which he purchased at the age of 25 (!). It’s transformation is described on Wikpedia thusly : “Observer has lost virtually all of its cultural currency among New York’s elite, but the paper is now profitable and reporting traffic growth” . Someone who can turn a newspaper profitable at the age of 25 in the year two thousand freaking ten should not be underestimated.

          * According to internal accounts, Trump gets to take the GOP voter info with him after the election. It is valued at >$100 million yearly, which would be a down-payment on the network (Russia Today’s budget is ~$300 million). Moreover, it would provide him with a direct, micro-targeted connection to specific partisan loyalists that’s unheard of for an up-start network. He would be able to reach out, in some form, to ~50 million registered republicans.

          Report

          • Those are solid points and he has Bannon on his side which is another plus. However starting up a network is a big deal. Fighting with Fox seems like a hard proposition and Trump has a short attention span. Maybe if its the crowd around him that is driving it letting him just have a place to rant with him not in control or leading. I still think a PPV website deal is more likely. If you have all that data on R’s you don’t need an entire tv network with all the other egos and struggles. But on the other hand ego is strong driver and stoking Trumpy’s ego is a monumental task. He may need millions of people just to maintain it.

            Report

      • Great bit of analysis from Nate. It’s shocking – SHOCKING – to find out that the margins of what’s viewed as a political sea-change (and rightly so) were so incredibly thin.

        Report

  10. Given that the states have some pretty different dynamics, such a prediction seems reasonable to me. (I have Dems winning the first two, though Missouri and Indiana are both odd in that Trump is running ahead of the GOP senate nominee.)

    Report

    • …though Missouri and Indiana are both odd in that Trump is running ahead of the GOP senate nominee.

      Don’t discount just how much Clinton is disliked in some parts of the country. In Colorado, Dem Senator Michael Bennett has consistently run well ahead of Clinton.

      Report

    • Bayh really saw his chances and stock plummet over the past week or so. Democrats seem divided on this. Some dislike Bayh and think good because they never forgave him for the “no labels” crap. Others while thinking Bayh was not liberal was as liberal as they could get from Indiana and argued he would be a loyal vote for the Ds most of the time.

      If HRC wins NC, I think the Democrats will get that Senate seat.

      Florida is interesting because Rubio is clearly favored but if there is a real Latino(a) surge in early voting, it is probably for HRC. The older generation of right-wing Cubans is going away. The newer Latino(a) votes tend to be more liberal second and third generation Cubans and liberal Puerto Ricans (who dislike the Cuban community.)

      Arizona is probably going to go McCain even if HRC carries the state. MO is also probably safe for Blunt.

      Report

    • My only disagreement with this statement is that there’s nothing odd about it at all. If I told you the Dems were expected to win 2 of those 5 seats and their chances at any of those 5 seats were individually the same, you wouldn’t expect them to be favored to win any of them, right?

      It’s an obviously flawed thought experiment because there’s probably some correlation between (say) an IN win and an MO win, but still.

      Report

      • IN and MO have some similarities, but in a populist change election, Bayh is a centrist-insider-DLC type, who worked in lobbying and private equity recently, while the Democratic challenger in MO served in Afghanistan and assembles assault weapons while challenging the old-establishment pol (who voted against border security) to show how fast he can assemble a weapon.

        Report

    • Are Senate Ds in odd position where they’re not favored in any single one of NC/MO/IN/FL/AZ but maybe favored to win at least one of them?

      Were the Indians in the odd position where they’re not favored in any single game, but it probably won’t be a sweep?

      Report

  11. Question for the crowd:

    How much split ticket voting do you think is going to happen? On the one hand, split ticket voting has been decreasing in successive election cycles. On the other hand, Trump and HRC are a unique dynamic.

    There are supposed to be a lot of new and/or first time voters in the Latino(a) surge. If this is true, they might be more likely to go straight ticket especially if they are turning up to dump Trump.

    Report

    • I think a lot.

      Generally speaking, the country is usually ready for a change after 8 years of one party in the White House. It’s one of the strengths of our system, IMHO. I think that dynamic is there this year, and that there were maybe 3 or 4 people the GOP could have nominated for POTUS that would have blown their chances — but they went ahead an nominated probably the worst of those 3 or 4.

      I expect a lot of moderates and independents will vote anti-Trump in the POTUS race, and anti-clinton in the Senate races.

      Report

      • Yeah, which is why I’m musing that maybe the Senate will be 51 Republican and 47 Democrats + 2 independents caucusing with them — a one-vote Republican majority.

        Because I just got another look at polling summaries from New Hampshire, a race about which I know very little other than a) both candidates have significant track records and b) what I read in polling summaries.

        Report

      • I expect a lot of moderates and independents will vote anti-Trump in the POTUS race, and anti-clinton in the Senate races.

        Colorado will undoubtedly be an outlier in this. Bennet was supposed to be the most vulnerable Dem Senate incumbent, until the state Republican party nominated a candidate who is quite possibly less competent than Trump.

        Darryl Glenn had commercials on during the Broncos game last night (after two-thirds of the votes have been cast), and honestly, the only visuals I can remember from them is the guy pumping iron and running a foothills trail all by himself. Nice set of muscles — makes Paul Ryan look kind of skinny.

        Report

  12. Clinton wins with 307 EV and with a plurality, not a majority. of the the popular vote. Republicans retain control of the Senate by one seat (51 seats). I previously believed the Senate would tie, but didn’t realize how poorly Bayh was running. Essentially I am predicting vote splitting in New Hampshire, but not Indiana or Missouri.

    Report

  13. I predict that whatever happens will be seen as obvious confirmation of everyone on every side’s notions about everything.

    Whatever it is you think, this election will confirm that you were right to think it.

    Report

  14. Clinton – 51.8 / Trump 43.9 / Johnson 3.6 / Stein 0.6

    Trump has no real GOTV operation and lots of married white women are Trump voters to their husbands, but Clinton voters in the ballot booth.

    Clinton 358 Trump 180 – All the swing states except GA go to Clinton, that is a narrow narrow Trump victory. Clinton comes within 5 or so in Texas.

    Senate – DNC 53 GOP 47 – GOP wins in AZ & FL – (Damn you Murphy) – Bayh wins by the seat of his pants. DNC wins all other tossups, including Missouri.

    House – GOP has 220 seats after the election. Biggest story of the Transition Period is whether or not the GOP can elect a speaker without Democratic help.

    Report

    • These all seem like pretty reasonable picks. By which I mean, close to my own (though a bit more bullish for the Dems).

      One thing that gives me a bit of pause is I’ve heard from more than one place that Trump’s GOTV is actually better than Romney’s was. Mostly because Romney tried to consolidate it and failed, whereas Trump simply handed everything over to the states, who will be able to do a better job than Romney’s operation did.

      One way to tell how true this is if Trump dramatically underperforms in Ohio compared to elsewhere, where the state GOP has more-or-less told him to go to hell.

      Report

      • I think the difference between Romney and Trump’s GOTV operation is this, from what I’ve read.

        Romney was basically trying to copy Obama’s system, but bluntly, he didn’t hire good people so it failed spectacularly in the run up to Election Day. But, on the other hand, Romney had the backing of the entire GOP infrastructure, so while there was no ‘secret sauce’, the normal GOP turnout machine worked.

        OTOH, with things being handed over totally to the states, in states with strong Republican parties, those parties will still be able to get the base, but they don’t have the extra technology to find those swing or unreliable voters the same way the Clinton machine can. But, on the other hand, you might see things tank where the state party is weak and more importantly, if the 55 year old white ladies who are the core of the GOP Election Day volunteer infrastructure much like 55 year old black ladies are the core of the DNC infrastructure have turned on Trump.

        For instance, I think Pennsylvania is going to be a spectacular disaster for Trump compared to the polls.

        Report

    • Question: Assuming the Dems get the Senate and assuming they make it abundantly clear that they intend to proceed with appointing Supreme Court justices one way or the other do you think the GOP will try and do their blockade with just the filibuster? Even knowing that ultimately it’ll get the filibuster wacked?

      Report

      • I don’t expect the incentives behind obstruction theater to disappear any time soon, so if this happens the filibuster will be abolished for SCOTUS appointments.

        Report

            • Yes, I thought for a while that he’d pull the nomination out of spite but that’s just silly. Of course he leaves the nomination out there until the very last day.

              The 115th Congress is seated on January 3, 2017; there will be 17 days between then and the swearing-in of POTUS No. 45. So the real interesting question is, could a Democrat-majority Senate, or possibly a 50-50 Senate with lame duck Vice President Biden, confirm Merrick Garland in seventeen days?

              Answer: yes, such a Senate could, IF its Republicans do not filibuster.

              Report

                • Right.

                  Garland was an “i’m obviously bending over backwards to give you a confirmable nominee” method to show just how cynical the Senate GOP is. He was perfect for that, but he’s far from the party’s dream justice (too old, to moderate).

                  I very much doubt Obama pushes for him post election, I’m sure Clinton wants her own person, and I don’t see who in the GOP caucus is willing to risk a primary challenge by raising the issue.

                  Report

                  • I’m actually interested to see how the threat of a primary challenge is going to play going forward. On the one hand, only one incumbent in the last two cycles has been displaced for being too moderate (and one for being too extreme). On the other hand, they don’t have a whole lot of air cover right now and nobody knows what to do.

                    As it pertains to Garland, I think it’s mostly a question of what McConnell wants to do.

                    Report

              • My impression is that, oddly enough, it only takes a majority to axe the fillibuster, but it’s something no majority has tried lest it backfire when they’re the minority. Hence, why it’s called the nuclear option.

                I could be completely wrong about the parlimentary procedure, though.

                Edit – also, that’s for fillibusters in the ‘not cloture’ practice they mostly do now – there’s nothing that can stop a team from doing the phone book reading thing if they have enough group discipine (which nobody has shown to possess these days)

                Report

                • Not quite.

                  The two issues in play are:

                  (1) Article 1, Section 5, Clause 2 of the Constitution authorizes each chamber of Congress to “determine the Rules of its Proceedings.”
                  (2) The reason the Senate has filibusters but the House doesn’t is that Aaron Burr screwed up in editing the Senate rules.

                  So the idea is that when the new Senate is formed, they could invoke the constitution to have whatever rules they want by majority vote. To date, they nearly always adopt the last Senate’s rules, but this time they could put whatever rules in they want (including those that would limite phone book reading filibusters, though I’m not aware of a movement to do that).

                  Report

                  • Remember that in the Senate there are two things: standing rules and precedents. Precedents are, fundamentally, clarifications or exceptions (usually narrow) to the standing rules or previous precedents. Precedents can be established by a simple majority. IIRC, there are some thousands of current precedents. The Senate Parliamentarian, who keeps it all straight, earns their money.

                    The filibuster exception for executive branch appointments and judges below the level of SCOTUS is a precedent. A SCOTUS exception would almost certainly be done as a precedent rather than a change to the standing rules.

                    Report

  15. For the sake of the polls, which seem to be converging, Clinton had better win by 4 points:

    Final national polls:NBC/SM: Clinton +6Ipsos: Clinton +4NBC/WSJ: Clinton +4ABC/WaPo: Clinton +4Herald: Clinton +4Bloomberg: Clinton +3— Matt McDermott (@mattmfm) November 7, 2016

    Report

      • They’re always off, but they’re usually more scattershot, so in a sence they’re not always off because “they” often aren’t saying any one thing. Rather, we’re forcing them to say one thing when they’re not via our (reasonable) practice of averaging them to get a picture of the race (however precise or casually it is that we’re doing the averaging).

        Here, it appears they are starting to actually come together to say one particular thing, more or less. I’d rather they be wrong through being ambiguous than be clear (convergent) like that, and wrong. At least, if the convergence is for a reason, rather than random. Clear and wrong, if it’s for a reason, means pollsters are flat-out doing it wrong. Scattershot and wrong just means there wasn’t a clear picture going in, it’s just that we were relying on the (good) theory that to average what there is to go on gives us the best picture that can be gotten from a group of varied results.

        Obviously, this depends on their being convergent for a reason, rather than purely randomly converging at the last moment before the election (which is obviously possible; probable probably.)

        Report

    • This is pretty much what I expect to happen. Clinton wins, by enough but no blowout.

      If Trump wins, it will blow my mind. His alt-right fanboys have already started saying there are “irregularities” in the vote, so we know they will say the whole thing was rigged (just like Daddy).

      Report

      • I think the think I’m most looking forward to on Wed morning is Scott Adams explaining to everyone how master persuaders know that losing the popular and the electoral vote is really a sign of Trump’s historic popularity, and how he now has Clinton exactly where he wants her.

        Report

          • I think the seeds of Adams’s current behaviour was there before but was kept in check by other factors. The problem with this political cycle is that he had one really good idea: that a lot of Trump’s success is due to an effective communication and sales strategy based on observable quirks in human cognition.

            The problem was, like a lot of people with one really good idea, he completely over-invested in its explainatory power, starting to think that all of politics and history can explained by this one organizational principle he’s come up with, rather than sensibly limiting himself to a good explaination of this one particular phenomena that was confunding people back in 2015. As he was doing this he appears to have ironically fallen prey to a wide range of cognitive fallacies that he typically calls other people out for, confirmation bias having pride of place as the most prominant one.

            I give him full marks for coming up with a good thesis for why Trump wasn’t just a random lucky idiot, but a guy with a certain skill set that gave him the opportunity to exploit political conditions ine America this presidential cycle. The problem was trying to make something more of it. Noticing the how Trump worked played to Scott Adams’ strengths in that he’s pretty clever and well versed in communication theory. The rest seems to be from his weaknesses of being wildly ignorant on most topics but still willing to give an opinion on them, that he routinely over-estimates his own cleverness and that he’s very comfortable being a jerk, thus doesn’t adjust well to people pointing out the flaws in his thinking.

            Basically he had both the strengths and weaknesses of being a clever but hugely overconfident and myopic middle-aged white guy. There was a lot of Dunning-Kruger going on. Not realizing that Trumps appeal to people of his demographic wouldn’t necessarily translate to others was a particularly big flaw in his thinking.

            Report

        • Scott Adams liked to sort of played that he was Dilbert and certainly his fans always always took him as the long suffering hero. It’s become clear Adams is sure he is Dogbert and Catbert but has far more pointy headed boss mixed in than he is aware of. The rest of us are just background characters.

          Report

        • It will be the masterstroke of Adam’s predictive persuasion magic-

          The linguistic kill shot will be that this is a rope a dope strategy by Trump who deliberately tossed the election to Hillary to totally neg America, until America decides it wants to surrender to Trump’s charm and prove how badly we want to make him a sammich.

          Report

  16. Pingback: Hillary Swings For The Swings | Hit Coffee

  17. On the eve of the election, I’ll stand by what I previously said.

    But it looks like I may have underestimated Trump a bit in Iowa, NC and FL. I think I’m wrong on Iowa, and if NC or FL flip, both will…

    So, Trump either gets 215 or 259.

    I’d like to see McMullin win Utah… not because I like McMullin (or Utah or winning) but because I like the idea of faction voting (or the legitimate threat of such) in a 2 party system. And that’s my counter-intuitive thought for the day.

    Report

Comments are closed.